Background
Fine particle annual mass concentrations in the Tennessee Valley range from 14 to20 micrograms per cubic meter. All seven urban/suburban sites exceeded the annual PM2.5 standard; only the rural Lawrence County TN site remained below the 15 µg/m3 annual standard. None of the stations exceeded the 65 µg/m3 level of the 24-hour PM2.5 standard. Summer high-winter low seasonality is evident. The current FRM PM2.5 mass measurements under-estimate the contribution of volatile/semi-volatile nitrates and organic carbon species. The semi-volatile organic fraction is both highly variable and significant, and assessments of semi-volatile and non-volatile organic carbon fractions are needed when particle composition measurements are made, especially at urban sites.

The U.S. Department of Energy's National Energy Technology Laboratory (DOE-NETL) and the Tennessee Valley Authority (TVA) are participating in an Interagency Agreement to collect and interpret air quality data in the Great Smoky Mountains National Park. This collaboration will provide for a better understanding of the relationship between coal-fired electric utility boiler emissions and PM2.5, ozone, and nitrogen loadings, the associated impact of these pollutants on the environment, and the need for future control strategies. The specific research under the TVA Interagency Agreement (IA) related to ambient monitoring activities are described below.

PM2.5 Sampling and Characterization
Rural and urban sites with extensive PM2.5 and co-pollutant monitoring capability are being installed and operated in the Southeastern United States to evaluate environmental and heath-related issues associated with ambient fine particulates. Monitoring sites include the Great Smoky Mountains National Park (GSMNP) (a rural site to evaluate visibility impacts) and Atlanta, Georgia (an urban site to evaluate ambient air concentrations and composition relative to health impacts). These research sites are equipped with various types of ambient monitors and filter media to allow for the collection of PM2.5 and other ambient air constituents for subsequent detailed chemical characterization. The sites are being operated in order to evaluate the precision and accuracy of mass data from the EPA monitoring network, to improve source-receptor and transport models, to assess diurnal, seasonal, and annual variations in composition, to develop sound management strategies; and to support epidemiological studies. The sites also serve as research platforms for testing new monitoring equipment and instrumentation.

To achieve these multiple objectives, a detailed understanding of the chemical composition of the ambient particulate matter is needed. At several of the field monitoring sites, a suite of advanced speciation equipment and instruments are being applied, including a semi-volatile sampler, a modified PM2.5 TEOM instrument, and a high-volume sampler for collecting samples for isotopic and trace species analysis. Every 3rd-day PM2.5 sampling at 8 sites in Tennessee and surrounding states was conducted using 3 prototype or FRM single-event fine particulate matter samplers. After gravimetric analysis, selected 24-hour samples were analyzed for elements Al to Pb using x-ray fluorescence (XRF) by EPA-approved Protocol 3, then extracted ultrasonically and analyzed for ammonium, sulfate and nitrate. For selected sampling days at the network's core sites, samples of fine mass were collected on collocated samplers using quartz as the collection medium. These quartz filters were analyzed by the thermo-optical reflectance (TOR) technique for organic and elemental carbon, as well as for ammonium, sulfate, and nitrate. Data were combined to deduce the average chemical composition of fine particles at the three core sites in the various seasons, with emphasis on compositional differences between urban and rural sites. More intensive winter and summer sampling was done at a mobile-source impacted site in

Chattanooga, with continuous measurements of mass, light scattering, and elemental carbon measurements (aethalometer). The data from these measurements were used to examine diurnal variations in mass and composition at this site and its potential effects on human exposure to fine particles. A new sampler (PC-BOSS) designed to accurately measure both non-volatile and semi-volatile constituents of fine mass was tested at urban and rural sites to test the accuracy of the prototype FRMs.

Ambient Sulfur Trends Analysis
As part of the analysis component of this task, investigations will be performed to resolve apparent contradictions among the results of various ambient air quality sampling programs performed to date. For example, data collected by the Clean Air Status and Trends Network (CASTNet) indicates that sulfur levels have declined in many areas of the eastern U.S. since about 1988. Much of this decline is thought to be due to decreases in sulfur dioxide (SO2) emissions from fossil-fuel-fired utility boilers as a result of Title IV reductions. However, some data collected under the Interagency Monitoring of Protected Visual Environments (IMPROVE) program suggest that PM2.5 sulfate levels have been increasing over parts of the southern Appalachian region. This apparent discrepancy needs to be investigated further.

In this project, a full year of PM2.5 data from the DOE/TVA/EPRI-funded PM2.5 Supersite and the immediately adjacent CASTNet and IMPROVE samplers on Look Rock being operated by the National Park Service (NPS) would be analyzed and compared. Since one of the possible explanations for the differences are the differences in the sampling frequency, the sampling frequency for the IMPROVE monitor would be increased from once every three days to every other day to enhance the comparability of the data from these three instruments. The resulting data from the enhanced sampling will be analyzed along with the existing data.

Nitrogen Deposition Monitoring - Noland Divide Watershed
Nitrogen deposition will also be closely examined to help determine what impact the new NOx emission controls (i.e., state regulations and NOx SIP call) have on environmental loadings of nitrogen compounds, including fine particulates. Nitrogen deposition monitoring will be carried out in the Noland Divide Watershed located in the high-elevation spruce-fir zone of the GSNMP. The study will investigate the linkages among nitrogen deposition, forest structure and dynamics, catchment topography, and stream water nitrogen exports. The results of this study will help to increase the understanding of the temporal, biogeochemical, and hydrologic links between atmospheric input and terrestrial and stream ecosystem concentrations.